Abstract
The pathophysiological mechanisms driving disease progression of frontotemporal lobar degeneration (FTLD) and corresponding biomarkers are not fully understood. Here we leveraged aptamer-based proteomics (>4,000 proteins) to identify dysregulated communities of co-expressed cerebrospinal fluid proteins in 116 adults carrying autosomal dominant FTLD mutations (C9orf72, GRN and MAPT) compared with 39 non-carrier controls. Network analysis identified 31 protein co-expression modules. Proteomic signatures of genetic FTLD clinical severity included increased abundance of RNA splicing (particularly in C9orf72 and GRN) and extracellular matrix (particularly in MAPT) modules, as well as decreased abundance of synaptic/neuronal and autophagy modules. The generalizability of genetic FTLD proteomic signatures was tested and confirmed in independent cohorts of (1) sporadic progressive supranuclear palsy-Richardson syndrome and (2) frontotemporal dementia spectrum clinical syndromes. Network-based proteomics hold promise for identifying replicable molecular pathways in adults living with FTLD. ‘Hub’ proteins driving co-expression of affected modules warrant further attention as candidate biomarkers and therapeutic targets.
Similar content being viewed by others
Main
Frontotemporal lobar degeneration (FTLD) is among the most common causes of early age of onset dementia (<65 years), resulting in progressive clinical syndromes that can feature behavioral, language and/or motor deficits1,2. The majority of FTLD comprises two major molecular subtypes, defined by pathological accumulation and cellular inclusions of hyperphosphorylated tau (FTLD-tau) or transactive response DNA-binding protein 43 kDa (FTLD-TDP)3. Current in vivo molecular biomarkers to aid in diagnosis of sporadic forms of FTLD are limited to nonspecific indicators of neurodegeneration and axonal loss (for example, neurofilament light chain (NfL)), as well as markers of Alzheimer’s disease (AD) pathology (for example, Aβ42/40 and P-tau181) to rule out AD. Although NfL predicts future clinical decline across FTLD clinical syndromes4,5,6, a single biomarker does not capture the biological complexity of FTLD and it remains unclear which biological pathways most strongly correlate with disease progression7. Molecular signatures that are detectable in vivo, inform underlying pathophysiology and contribute to clinical progression are urgently needed to advance understanding of FTLD pathogenesis and guide biomarker development, which will in turn enhance FTLD clinical trials through improved selection of patients and measurement of treatment effects8.
The underlying FTLD pathology can be confidently predicted in inherited cases caused by autosomal dominant mutations, which represent approximately 20–30% of all cases9. Observational study of FTLD mutation carriers has provided a unique opportunity to chart clinical and biomarker changes across presymptomatic and symptomatic stages of FTLD5,8,10. Autosomal dominant forms of FTLD-TDP are most often caused by mutations in chromosome 9 open reading frame 72 (C9orf72) or progranulin (GRN). The majority of autosomal dominant FTLD-tau is due to mutations in microtubule-associated protein tau (MAPT)5,9. Current understanding of molecular mechanisms underlying autosomal dominant FTLD has been primarily ascertained from postmortem and non-human studies3,11. Hexanucleotide repeat expansions in the C9orf72 gene may produce both loss-of-function (haploinsufficiency) and toxic gain-of-function (for example, repeat RNA foci and dipeptide repeat protein aggregates) mechanisms that impair RNA metabolism, nucleocytoplasmic transport and proteostasis pathways that contribute to nuclear depletion and cytoplasmic aggregation of TDP-43 (refs. 12,13). GRN mutations cause haploinsufficiency of progranulin, a lysosomal protein that regulates immune and autophagy pathways implicated in TDP-43 dyshomeostasis14,15. MAPT mutations alter tau binding to microtubules and promote tau aggregation, which directly compromises axonal transport, synaptic signaling, cellular metabolism and the extracellular scaffolds that support neuronal structures16,17,18.
Recent large-scale proteomics studies in AD are contributing to discovery of key biological networks and individual proteins reflective of AD pathophysiology, which extend far beyond classic hallmarks of misfolded amyloid and tau19,20. These studies have employed powerful systems-biology approaches in brain tissue and biofluids (for example, cerebrospinal fluid (CSF) and blood) to identify communities of co-expressed proteins that are reliably altered in clinical cohorts that span the AD disease severity continuum20,21,22. Despite potential to identify molecular signatures underlying disease progression, unbiased large-scale analysis of the biofluid proteome in living patients with FTLD has lagged.
The present study employed a large-scale aptamer-based proteomic approach (SomaScan) to generate a CSF protein network in 116 participants with autosomal dominant FTLD mutations (C9orf72, GRN and MAPT). Network analysis identified dysregulated communities of CSF proteins (modules) involved in RNA splicing, protein degradation, extracellular matrix (ECM) and synaptic pathways. These modules were strongly correlated with cross-sectional and longitudinal markers of FTLD clinical progression. We validated the generalizability of our genetic FTLD protein network by demonstrating preservation of network module alterations in CSF of progressive supranuclear palsy (PSP), a sporadic four-repeat FTLD tauopathy. We further evaluated the disease specificity and biologic reproducibility of protein network associations by reconstructing modules in a third clinical cohort of sporadic and genetic FTLD and AD using an orthogonal proximity-extension assay-based proteomic approach (Olink), as well as cross-referencing FTLD signatures with proteomic datasets in AD and Parkinson’s disease (PD). Collectively, our approach and results provide insights regarding in vivo molecular signatures of central nervous system dysfunction that occur across the disease severity spectrum of FTLD.
Results
Participant characteristics
Supplementary Table 1 reports the demographic and clinical characteristics for the 116 autosomal dominant FTLD mutation carriers and 39 familial controls, representing cognitively normal non-mutation carriers from families with a known C9orf72, GRN or MAPT mutation. Consistent with the broader ARTFL/LEFFTDS Longitudinal Frontotemporal Lobar Degeneration (ALLFTD) cohort10, C9orf72 was the most common mutation (N = 47), followed by MAPT (N = 37) and GRN (N = 32). At baseline, almost half of all mutation carriers were presymptomatic (global Clinical Dementia Rating Scale (CDR) plus Behavioral and Language Domains from the National Alzheimer’s Coordinating Center (NACC) FTLD module of 0; 54 (47%); 24 (51%) C9orf72, 12 (39%) GRN and 18 (49%) MAPT). GRN carriers were older on average (mean age 57.9 years) than C9orf72 (mean age 49.9 years), MAPT (mean age 44.5 years) and controls (mean age 45.8 years). Mutation carriers and controls were comparable in terms of sex (mutation carriers: 49% to 55% female versus controls: 59% female) and education (mutation carriers: mean 15.0–15.7 years versus controls: mean 15.6 years). The median (interquartile range) of annual study visits was 3 (1.75–4) for mutation carriers and 3 (2–5) for controls.
Genetic FTLD protein co-expression network
The study design is shown in Fig. 1. Using weighted gene correlational network analysis (WGCNA), we built a protein co-expression network from 4,138 proteins across 155 CSF samples to identify protein communities that are dysregulated in FTLD (Fig. 2a). Network analysis revealed 31 protein co-expression modules across controls and mutation carriers. Module size ranged from 360 proteins (M1) to 48 proteins (M31). Protein module membership assignments are provided in Supplementary Table 2. On the basis of Gene Ontology (GO) and cell type enrichment analyses (Supplementary Table 3 and Supplementary Fig. 1), 28 out of the 31 modules had a clear primary ontology that was used for annotation, while 3 modules did not evidence clear ontology (labeled ‘ambiguous’).
CSF was collected in 116 carriers of autosomal dominant mutations for FTLD (47 C9orf72, 32 GRN and 37 MAPT) and 39 non-carrier controls with a family history of genetic FTLD. CSF was analyzed on a modified aptamer-based assay (SomaScan). After data processing, a total of 4,138 proteins were quantified. High-dimensional proteomic data were organized into modules of protein co-expression using WGCNA. CSF protein co-expression modules from the genetic FTLD network were functionally annotated using gene set and cell type enrichment approaches. CSF genetic FTLD modules were examined in relation to cross-sectional (CDR + NACC-FTLD, CSF NfL and bilateral frontotemporal volume) and longitudinal (global cognitive trajectories) indicators of disease severity. On cross-cohort and cross-platform validation analyses, genetic FTLD modules were reconstructed in independent cohorts of sporadic PSP-RS and controls (4RTNI; SomaScan) and FTLD, AD and controls (BioFINDER 2; Olink). Figure created with BioRender.
a, A CSF protein co-expression network was built using WGCNA. The genetic FLTD network consisted of 31 protein co-expression modules. Module relatedness is shown in the dendrogram (right). GO analysis was used to identify the principal biology represented by each module. Within genes, module eigenproteins in symptomatic (Sx) and presymptomatic (PreSx) variant carriers were compared against controls. Increased eigenprotein abundance in FTLD is indicated in green, whereas decreased eigenprotein abundance is indicated in blue. Module eigenproteins were correlated with disease outcomes, including CDR + NACC-FTLD, global cognitive slope, CSF NfL and bilateral frontotemporal volumes (red, positive correlation; blue, negative correlation). The cell type nature of each module was assessed by module protein overlap with cell type-specific marker lists of neurons, oligodendrocytes, astrocytes, microglia and endothelia. The asterisks in the left heat map indicate statistical significance after one-way ANOVA with Tukey’s test (two tailed) ***Tukey P < 0.001, **Tukey P < 0.01 and *Tukey P < 0.05. The asterisks in the middle and right heat maps indicate statistical significance after FDR correction: ***FDR P < 0.001, **FDR P < 0.01 and *FDR P < 0.05. The exact P values are reported in Supplementary Table 4. b, Module eigenprotein levels by case status for six of the most strongly FTLD-associated modules. Individual eigenprotein values are plotted in controls (N = 39) and mutation carriers, which were grouped by Sx and PreSx status (C9orf72: 24 PreSx, 23 Sx; GRN: 12 PreSx, 19 Sx; MAPT: 18 PreSx, 19 Sx). Red P values are statistically significant. Differences in module eigenprotein by case status were assessed by one-way ANOVA with Tukey’s test. Gene-specific P values represent the omnibus significance for gene-stratified comparisons versus controls, with significant effects displayed with red text. The box plots represent the median and 25th and 75th percentiles, and box hinges represent the interquartile range of the two middle quartiles within a group. Min and max data points define the extent of whiskers (error bars). C9, C9orf72; CTL, control.
CSF proteomic signatures of FTLD disease severity
Analyses first combined FTLD mutation carriers across all three genes (N = 116) to maximize sample size. Module eigenproteins, calculated as the first principal component of module abundance, were compared across symptomatic carriers, presymptomatic carriers and controls using one-way analysis of variance (ANOVA) with Tukey’s post hoc correction, adjusting for age and sex. Eigenproteins were also correlated with continuous indicators of functional severity (CDR + NACC-FTLD sum of boxes), adjusting P values with false discovery rate (FDR) corrections according to the Benjamini–Hochberg method. Seven of 31 modules showed significant differences between symptomatic carriers and controls (Supplementary Table 4). Four modules exhibited particularly strong relationships with functional severity: M26 spliceosome, M2 presynapse, M28 synapse assembly/axon and M22 autophagy. M26 spliceosome, highly enriched for proteins involved in mRNA splicing and nuclear transport, was markedly increased in symptomatic carriers versus controls and positively correlated with functional severity (ρ = 0.41, FDR P = 2.7 × 10−6). M2 presynapse and M28 synapse assembly/axon, enriched for synaptic/neuronal (M2 and M28) and oligodendrocyte (M28) cell-type markers, were decreased in symptomatic carriers versus controls and negatively correlated with functional severity (M2: ρ = −0.33, FDR P = 2.6 × 10−4; M28: ρ = −0.35, FDR P = 1.7 × 10−4). M22 autophagy was also decreased in symptomatic carriers versus controls and negatively correlated with functional severity (ρ = −0.33, FDR P = 2.6 × 10−4). CSF proteomic alterations were more subtle and less likely to survive post hoc correction when comparing presymptomatic mutation carriers versus controls. However, M9 ion transport, enriched for neuronal cell-type markers involved in ion channel activity and transport, was significantly decreased in presymptomatic carriers versus controls.
We next examined how network modules associated with frontotemporal brain volumes and CSF NfL, the most established fluid biomarker for disease monitoring in genetic and sporadic FTLD4,23,24. Ten modules were associated with frontotemporal volumes after FDR correction. Modules that negatively correlated with symptom severity were positively correlated with brain volumes, and vice versa, including M2 presynapse (ρ = 0.40, FDR P = 1.3 × 10−5), M28 synapse assembly/axon (ρ = 0.36, FDR P = 1.1 × 10−4), M22 autophagy (ρ = 0.32, FDR P = 9.6 × 10−4) and M26 spliceosome (ρ = −0.24, FDR P = 1.7 × 10−2). Six modules associated with CSF NfL concentrations were measured by Simoa at an unadjusted P < 0.05, with only M26 spliceosome surviving FDR correction (ρ = 0.32, FDR P = 9.3 × 10−3). Upon closer examination, we observed that the SomaScan aptamer for NfL (NEFL) was assigned to M26 spliceosome. However, NEFL was not a strong driver of M26 eigenprotein co-expression (intramodular connectivity (kME) of 0.32 (87th out of 88 module members)) and the relationship between M26 spliceosome and CSF NfL (Simoa) persisted when recalculating the M26 spliceosome eigenprotein without NEFL (ρ = 0.31, P = 9.1 × 10−3).
Gene-stratified FTLD proteomic signatures
We performed gene-stratified analyses to determine whether proteomic signals observed in the full sample were driven by specific gene groups. For each gene group, we again compared module eigenprotein levels across symptomatic mutation carriers, presymptomatic mutation carriers and controls using Tukey’s post hoc correction. We also examined correlations with CDR + NACC-FTLD sum of boxes, CSF NfL and frontotemporal volumes (Fig. 2b and Supplementary Table 4). FDR corrections were not applied to gene-stratified correlational analyses given reduced sample sizes. Within each gene group, M26 spliceosome levels were increased and M22 autophagy module levels were decreased in symptomatic carriers versus controls. However, symptomatic C9orf72 and GRN carriers exhibited more pronounced alterations in M26 and M22 (0.85–1 difference in z score versus controls) than symptomatic MAPT (0.5–0.6 difference in z score versus controls). Presymptomatic GRN carriers also showed elevations in M26 spliceosome that were not present in presymptomatic C9orf72 or MAPT mutation carriers. M2 presynapse and M28 synapse assembly/axon were decreased in symptomatic GRN and MAPT, with both modules exhibiting particularly strong relationships with CDR + NACC-FTLD and CSF NfL in MAPT (M2: CDR + NACC-FTLD ρ = −0.59, P = 1.0 × 10−4; NfL ρ = −0.54, P = 8.0 × 10−4; frontotemporal volume: ρ = 0.55, P = 1.1 × 10−3; M28: CDR + NACC-FTLD ρ = −0.43, P = 7.4 × 10−3; NfL ρ = −0.46, P = 5.0 × 10−3; frontotemporal volume: ρ = 0.30, P = 9.7 × 10−2). M9 ion transport, which showed the strongest presymptomatic signal in full sample analyses, was decreased among both presymptomatic C9orf72 and MAPT carriers versus controls.
Several additional module expression patterns emerged in gene-stratified analyses that were not present when examining modules in all mutation carriers. A cluster of modules adjacent to M26 spliceosome that also exhibited GO enrichment for ‘RNA binding’ (M24 ubiquitination/translation, M25 protein folding/metabolism and M27 metabolism) were increased and/or positively correlated with disease severity in both symptomatic C9orf72 and GRN. M29 ECM, enriched for ECM proteins and microglial cell-type markers, was uniquely elevated in symptomatic MAPT carriers versus controls. M4 complement/coagulation was closely related to M29 and also selectively elevated in symptomatic MAPT versus controls. M3 postsynapse/glycosylation did not significantly differ between mutation carrier groups and controls, but was one of the strongest correlates of disease severity measures in MAPT (CDR + NACC-FTLD ρ = −0.53, P = 7.0 × 10−4; NfL ρ = −0.46, P = 5.9 × 10−3; frontotemporal volume: ρ = 0.53, P = 1.6 × 10−3).
Differential abundance and ROC analysis
To complement network analyses, we examined differential abundance of all 4,138 proteins to identify individual CSF proteins driving module-level differences in gene-stratified comparisons of symptomatic carriers, presymptomatic carriers and controls (Supplementary Table 5 and Extended Data Fig. 1). M26 spliceosome had the highest proportion of individual differentially abundant proteins in symptomatic C9orf72 and GRN carriers (Supplementary Table 6), including the nuclear proteins TRA2B, TMPO and HNRNPAB. Proteins assigned to modules with neuronal enrichment were also differentially abundant across multiple gene groups versus controls, including established synaptic markers NPTX2 (M2 presynapse), CNTNAP2 (M28 synapse assembly/axon), DLG4 (M3 postsynapse/glycosylation) and 14-3-3 proteins (for example, YWHAZ, M19 neuron migration). Symptomatic MAPT exhibited the highest proportion of differentially abundant proteins from these neuronal modules, particularly M28 synapse assembly/axon. M26 spliceosome had the highest proportion of differentially abundant protein members in presymptomatic GRN (for example, XRCC6) and M9 ion transport had the highest proportion of differentially abundant protein members in presymptomatic C9orf72 and MAPT (for example, KCNE2 and GPR6).
GRN mutations are characterized by progranulin protein haploinsufficiency. As expected, CSF progranulin protein (unassigned to a module) exhibited the largest decreased abundance in both pre- and symptomatic GRN carriers versus controls. C1QTNF1 (unassigned to a module), another lysosomal protein with CSF levels shown to be strongly colinear with GRN levels25, exhibited a similar pattern of decreased abundance in GRN carriers.
We next performed receiver-operating characteristic (ROC) analyses to evaluate discriminative ability of proteins with differential abundance across FTLD group comparisons. The top five performing proteins for each gene-stratified group comparison reached an area under the curve (AUC) value of 0.77 or higher (Extended Data Fig. 2), with strongest separation occurring between symptomatic carriers versus controls (C9orf72 top hit: TMPO (AUC of 0.93); GRN top hit: GRN (AUC of 0.97); MAPT top hit: NPTX2 (AUC of 0.88)). In combination, a top five protein panel for each comparison reached an AUC value of 0.88 or higher, with most comparisons surpassing 0.90. Supplementary Table 7 reports AUC values with 95% confidence intervals and sensitivity/specificity thresholds for all differentially abundant proteins across group comparisons.
Module associations with cognitive trajectories
Module correlations with global cognitive trajectories in the full sample are provided in Fig. 3 and Supplementary Table 7. Modules that were most strongly associated with cognitive decline in the full sample included M29 ECM (ρ = −0.39, FDR P = 2.1 × 10−5) and M26 spliceosome (ρ = −0.22, FDR P = 3.9 × 10−2). Modules that were most strongly associated with cognitive preservation included M28 synapse assembly/axon (ρ = 0.42, FDR P = 7.8 × 10−6), M2 presynapse (ρ = 0.41, FDR P = 7.8 × 10−6) and M3 postsynapse/glycosylation (ρ = 0.33, FDR P = 5.6 × 10−4).
The plots display the top five CSF genetic FTLD network modules most strongly associated with global cognitive trajectories in the full sample. Eigenprotein z scores are plotted against the annual rate of global cognitive change during the study period (n = 137). Person-specific cognitive slopes were extracted from linear mixed-effects models that included baseline demographics (age, sex and education) and time (years since baseline). Regression fit lines with 95% confidence intervals are plotted alongside Spearman’s ρ values and two-tailed P values. Information on the association between all network module eigenproteins and cognitive trajectory in the full sample, within each gene and within presymptomatic mutation carriers is provided in Supplementary Table 8. For proteins assigned to each module, an individual protein’s strength of connectivity to the module (x axis) is plotted against the individual protein’s correlation with global cognitive change (y axis). Proteins that exhibited stronger intramodular connectivity also exhibited stronger relationships with cognitive slope. Color-filled triangles represent individual proteins that survived FDR correction in two-tailed proteome-wide differential correlational analyses (n = 646 total proteins; full list in Supplementary Table 8). Proteins in the top 20th percentile of intramodular connectivity are classified as ‘hub’ proteins. Hub proteins that significantly correlated with cognitive trajectory are listed with each plot.
To determine whether cognitive findings were driven by specific gene groups, post hoc analyses examined gene-stratified correlations between modules and cognitive trajectories (Supplementary Table 8). M26 spliceosome exhibited one of the strongest associations with cognitive trajectories in GRN, whereas M29 ECM and synaptic/neuronal modules (M2 presynapse, M3 postsynapse/glycosylation and M28 synapse assembly/axon) were among the strongest contributors to cognitive trajectories in C9orf72 (M29, M2 and M28) and MAPT (M29, M2 and M3). To determine the relationships between CSF modules and early stage cognitive change, analyses were also conducted within presymptomatic carriers, collapsed across gene group. In these analyses, M29 ECM (ρ = −0.44, P = 1.8 × 10−3) and M2 presynapse (ρ = 0.44, P = 1.7 × 10−3) were most strongly associated with early cognitive change.
On differential correlational analyses, 646 proteins were significantly correlated with global cognitive trajectories in the full sample (FDR P < 0.05; Supplementary Table 8). Over half of those differentially expressed proteins were assigned to modules that also strongly associated with cognitive trajectories (M26, M29, M2, M3 and M28). To determine whether individual proteins linked to cognitive trajectories were also of high influence (‘hub proteins’) in these target modules, we plotted individual protein correlations with cognitive slope against their intramodular kME. Proteins that were module hubs, defined as being within the top 20th percentile of kME for their respective module20, and had FDR-corrected significance with cognitive slope are highlighted in Fig. 3 inset protein lists. Notable ‘hub’ proteins included neuronal pentraxins, NPTX2 (M2 presynapse; largest effect on cognitive slope across all proteins) and NPTX1 (M28 synapse assembly/axon), as well as neuroligins NLGN1 and NLGN2 (M3 postsynapse/glycosylation). Other notable ‘hub’ proteins linked to cognitive trajectories included transmembrane proteins TMEM106B (M3 postsynapse/glycosylation) and TMEM132B (M2 presynapse), ECM-linked proteins FSTL1 and TIMP1 (M29 ECM) and nuclear proteins HNRNPA1 and RECQL (M26 spliceosome).
Genetic FTLD modules are preserved in sporadic PSP-RS
We applied the same WGCNA methods to CSF SomaScan data from an independent cohort of individuals with sporadic PSP-Richardson syndrome (RS)26 to determine whether protein co-expression patterns identified in genetic FTLD could be reproduced in sporadic FTLD, specifically a sporadic FTLD tauopathy (Supplementary Table 9a). All 31 modules from the genetic FTLD cohort were highly preserved in the sporadic PSP-RS network (all zsummary scores >10 (>q = 1 × 10−23)), indicating strong consistency of the CSF proteomic correlational architecture across genetic and sporadic disease (Fig. 4a). Next, we reconstructed genetic FTLD network modules in PSP-RS using synthetic eigenproteins to determine whether modules derived from the genetic network could also differentiate PSP-RS from controls. Of the 31 synthetic eigenproteins derived from the genetic FTLD protein module assignments, five were significantly altered in sporadic PSP-RS versus controls after FDR correction (decreased in PSP-RS: M28 synapse assembly/axon, M2 presynapse, M3 postsynapse/glycosylation and M19 neuron migration; increased in PSP-RS: M29 ECM; Fig. 4b and Supplementary Table 9b). Notably, four of these five modules were those most strongly related to MAPT disease severity (M2, M3, M28 and M29), with consistency in directionality of effects (Fig. 4c). These results suggest that, despite differences in initial pathogenesis, genetic and sporadic forms of FTLD-tau exhibit shared proteomic signatures in CSF, characterized by decreases in neuronal cell type-specific proteins and increases in ECM proteins.
Validation cohorts included 4RTNI, composed of PSP and controls, and BioFINDER 2, composed of patients with FTLD clinical syndromes, biomarker-confirmed AD and controls. The 4RTNI CSF samples were assayed with SomaScan and BioFINDER CSF samples were assayed with Olink. a, WGCNA was applied to 4RTNI SomaScan data to test for module preservation across the genetic FTLD and sporadic PSP-RS networks. Modules that have a zzsummary score greater than or equal to 1.96 (or q =0.05, dotted blue line) are considered to be preserved, and modules that have a zsummary score greater than or equal to 10 (or q = 1 × 10−23, dotted red line) are considered to be highly preserved. All modules in the genetic FTLD network were highly preserved in the sporadic PSP-RS network. b, Synthetic eigenproteins were reconstructed in 4RTNI and BioFINDER 2 to test for concordance in module relationships with disease groups. The heat map displays average synthetic eigenprotein z score differences between PSP-RS and controls (CTL), FTLD versus CTL, FTLD versus AD and AD versus CTL. The asterisks indicate statistical significance after pairwise two-sided t-test with FDR correction. ***FDR P < 0.001, **FDR P < 0.01 and *FDR P < 0.05. The exact P values are reported in Supplementary Tables 9 and 10. c,d, Synthetic eigenprotein box plots for key modules from 4RTNI (35 CTL and 30 PSP-RS) (c) and BioFINDER 2 (248 CTL, 58 AD and 29 FTLD) (d) analyses. Pairwise differences in module synthetic eigenproteins by case status were assessed by two-sided t-tests with FDR correction. The box plots represent the median and 25th and 75th percentiles, and box hinges represent the interquartile range of the two middle quartiles within a group. Min and max data points define the extent of whiskers (error bars).
The CSF proteome differentiates FTLD from other diseases
To assess whether proteomic signatures from the FTLD network were specific to FTLD or more broadly reflective of neurodegeneration, irrespective of molecular etiology, we constructed synthetic eigenproteins in a second replication cohort (BioFINDER 2; Supplementary Table 10a) comprising 29 patients with frontotemporal dementia clinical syndromes, 87 AD patients matched on demographics and disease severity, and 248 AD biomarker-negative controls. BioFINDER 2 CSF samples were analyzed using a proximity-extension assay platform (Olink). Synthetic eigenproteins were computed based on protein measurements that overlapped between Olink and SomaScan, offering the opportunity to validate proteomic signatures across platforms. CSF proteomic alterations were more pronounced in FTLD than AD. Thirteen synthetic eigenproteins differed between FTLD and controls, six differed between FTLD and AD and only two differed between AD and controls (all FDR P < 0.05; Supplementary Table 10b). Most synthetic eigenproteins that differentiated FTLD from AD and/or controls represented modules that also differed between FTLD mutation carriers and familial controls from ALLFTD (Fig. 4d). These included increased M26 spliceosome (FTLD > AD > control) and decreased M2 presynapse (FTLD < AD and control), M28 synapse assembly/axon (FTLD and AD < control) and M22 autophagy (FTLD < AD and control).
In addition to cross-platform synthetic module analyses, we tested specificity of the top CSF SomaScan protein hits to FTLD by correlating gene-stratified symptomatic FTLD-associated protein alterations with CSF SomaScan protein changes reported by prior studies in AD27 and PD28. We observed significant but small correlations between FTLD-associated proteins and AD-associated proteins (Pearson r range of 0.14–0.21; Extended Data Fig. 3a,c,e). We observed medium-sized correlations between FTLD-associated and PD-associated proteins (r range of 0.46–0.56; Extended Data Fig. 3b,d,f), yet smaller than the large correlations reported by prior studies that have compared effect sizes for the same disease across the same platform27,29. The stronger concordance between FTLD and PD was driven by shared decreases in neuronal markers of neurodegeneration from M2 presynapse (for example, NPTX2, LRFN2 and NRXN1) and M22 autophagy (for example, SEMA4D), whereas these modules were enriched for proteins with increased abundance in AD (Extended Data Fig. 3g and Supplementary Table 11). M26 spliceosome lacked enrichment for AD- and PD-associated proteins, as many key members of M26 (for example, HNRNPA2B1, TMPO, ATP5J and RECQL) were selectively increased in FTLD but not AD or PD. Collectively our disease-specificity validation analyses support the cross-platform and cross-cohort reproducibility of FTLD network-derived modules as well as unique associations of modules and individual targets with FTLD compared with other neurodegenerative diseases.
Module overlap with CSF AD networks
Protein co-expression patterns and pathway enrichment present in the CSF genetic FTLD network partially resembled CSF networks previously identified in AD21,27,30. We employed module overrepresentation analyses to empirically compare the genetic FTLD network with previously constructed CSF networks in sporadic AD and thus identify CSF modules that overlap across neurodegenerative conditions. These AD networks were built using multiplatform (SomaScan + tandem mass tag-based mass spectrometry (TMT-MS); Dammer et al.27) and TMT-MS (Modeste et al.30) proteomic approaches, which also allowed us to probe the influence of platform on FTLD and AD network overlap (Fig. 5 and Supplementary Table 12). Of the 31 genetic FTLD network modules, 21 had significant overrepresentation of protein members in at least one corresponding module from the multiplatform AD network (31 of 34 modules) and 11 had significant overrepresentation in at least one module from the TMT-MS AD network (10 of 14 modules). M4 complement/coagulation was the most strongly overlapping FTLD network module in both AD networks, which also exhibited the strongest preservation across brain tissue, CSF and plasma in multiplatform studies21,27. M29 ECM and neuron/oligodendrocyte-enriched FTLD network modules (M2, M3, M28 and M31) also exhibited strong overlap with modules from both CSF AD networks. M26 spliceosome most strongly overlapped with M14 translation from the multiplatform network, which did not differ between AD and controls. M26 spliceosome did not overlap with any module from the TMT-MS network, which broadly lacked modules with primary enrichment for RNA binding/splicing pathways. Overall, we observed strong conservation of modules linked to neuronal, oligodendrocyte, ECM and immune processes in both FTLD and AD CSF samples.
Module member ORA of the genetic FTLD CSF network with two CSF protein co-expression networks in AD. Top: a multiplatform network, obtained using SomaScan and TMT-MS (Dammer et al.27). Bottom: a single platform network obtained using TMT-MS (Modeste et al.30). The box values represent −log10(FDR) value for pairwise module overlap, determined using one-tailed Fisher’s exact test. Bolded AD network modules significantly differed between AD and controls. Modules from the AD networks (y-axis rows) without an overlap value of −log10(FDR) >1 are not included in the heat maps. ***FDR P < 0.005, **FDR P < 0.01 and *FDR P < 0.05. The exact P values are reported in Supplementary Table 12.
Discussion
We analyzed over 4,000 CSF proteins from patients across the disease severity spectrum of genetic FTLD. Protein co-expression communities most strongly linked to FTLD included proteins involved in RNA processing, synaptic/axonal function, ECM and protein degradation. These modules, and hub proteins that drive their co-expression, correlated with cross-sectional markers of clinical severity and axonal degeneration as well as longitudinal cognitive change. We also observed CSF protein alterations in presymptomatic mutation carriers, including decreased M9 ion transport in C9orf72 and MAPT, and increased M26 spliceosome in GRN. Genetic FTLD proteomic signatures were also reproduced in patients with PSP-RS with autopsy-confirmed PSP (sporadic FTLD-tau), whose network alterations closely resembled the neuronal and ECM-linked signatures of MAPT carriers (genetic FTLD-tau). Genetic FTLD-associated modules also distinguished a cohort comprising both sporadic and genetic frontotemporal dementia patients from controls and AD, despite cross-cohort differences in proteomic measurement platforms. Our findings support the utility of large-scale CSF proteomics in identifying biological pathways and candidate biomarkers across the spectrum of FTLD disease severity, with relevance to both genetic and sporadic forms of disease.
Spliceosomal and related nuclear proteins were markedly increased in symptomatic carriers across all three FTLD gene groups. Prior studies report increased RNA splicing protein levels across neurodegenerative disease brain tissues, including FTLD-TDP31 and tauopathies32, implicating RNA splicing dysfunction as a shared mechanism of neurodegeneration33. Cross-cohort replication analyses in the BioFINDER 2 Olink dataset demonstrated a stair-step pattern of increased M26 spliceosome abundance across controls, AD and FTLD. Although M26 overlapped with RNA pathway modules from the multiplatform AD network27, many FTLD-associated proteins from M26 did not evidence a similar signal in the AD network. This suggests that CSF-detectable spliceosome alterations may be more pronounced in FTLD, with distinct contributions to its pathophysiology compared with AD. M26 was elevated before symptom onset and predicted cognitive decline in GRN carriers, highlighting the early stage prognostic utility of spliceosomal proteins in genetic FTLD-TDP. It is possible that increased CSF levels of RNA splicing proteins in FTLD-TDP may reflect compensatory or pathogenic protein network alterations in response to nuclear depletion of TDP-43 and related deficits in RNA metabolism and nucleocytoplasmic transport34,35. Nuclear depletion of TDP-43 increases production of HNRNP proteins36,37, which were also differentially abundant hub proteins in M26 spliceosome (for example, HNRNPA1, HNRNPA2B1 and HNRNPAB). We also detected increased M26 spliceosome abundance in symptomatic MAPT carriers, suggesting relevance to primary tauopathies though probably through mechanisms divergent to those observed in TDP-43. Although speculative, it has been proposed that tau regulates aspects of RNA physiology under healthy conditions and mutant tau aggregates may sequester critical RNA splicing factors binding proteins into the cytoplasm38. As efforts continue to develop diagnostic biofluid markers that measure the highly specific consequences of TDP-43 loss of function39 and primary tauopathies40, orthogonal development of targeted assays for proteins involved in nuclear processing may be valuable for prognosis, disease monitoring and measuring therapeutic response.
Modules enriched for neuron-derived proteins, including M2 presynapse, M28 synapse assembly/axon and M3 postsynapse/glycosylation, were strong predictors of cognitive decline. This neuronal signature of FTLD-related cognitive decline aligns with prior studies demonstrating that synaptic integrity is critical for cognitive function41,42,43. A decrease in synaptic protein abundance in FTLD, which was observed across all three study cohorts, may reflect alterations in synaptic protein turnover due to decreased protein production44. CSF NPTX2, previously shown to decrease in symptomatic FTLD and other neurodegenerative diseases45,46,47, exhibited the strongest association with cognitive decline out of any individual protein. Combining NPTX2 with NfL and additional synaptic markers of neurodegeneration may enhance the molecular phenotyping of neurodegeneration in FTLD and help identify patients at highest risk of cognitive progression.
Decreased abundance of M9 ion transport was a striking feature of both presymptomatic C9orf72 and MAPT. M9 was enriched for proteins that govern neuronal ion flux, including potassium voltage-gated channel proteins (for example, KCNIP4 and KCNE2) that regulate neuronal excitability and contain genetic risk loci for psychiatric disorders48,49. Presymptomatic alterations to M9 may reflect neuronal ion imbalances, which could contribute to neural signaling changes50,51 that often precede widespread neurodegeneration in FTLD52. Decreased M9 ion transport levels may also reflect a longstanding neurodevelopmental molecular signature of genetic FTLD, considering that M9 levels did not correlate with clinical severity. M9 hub proteins such as NNAT and FOXG1 are key regulators of early stage neuron differentiation53,54. Longitudinal proteomic profiling will help elucidate the timing and trajectory of changes in these neurodevelopmental proteins, which will inform their suitability as early stage FTLD biomarkers.
Similar to synaptic modules, the microglial-enriched M29 ECM predicted cognitive trajectories, even among presymptomatic mutation carriers. M29 hub proteins included ECM-secreted growth factors (for example, FSTL1 and PDGFD) as well as lysosomal and angiogenic proteins that mediate protein degradation and remodeling of the ECM (for example, CTSD and TIMP1). In addition to providing structural support for neurons, the ECM also modulates intercellular communication and synaptic plasticity55,56. In AD brains, ECM proteins are positively correlated with tau pathology20 and ECM glycoproteins are elevated in CSF of autosomal dominant AD several decades before symptom onset57, suggesting that ECM dysregulation is an early event in AD pathogenesis. M29 was more strongly correlated with symptomatic MAPT and PSP-RS than C9orf72 or GRN, which could reflect interactions between the ECM and extracellular levels of tau that are released during trans-cellular spread58,59.
Although our study sample is large for a CSF proteomics study in FTLD, it is smaller than CSF proteomics studies from more prevalent neurodegenerative conditions. Our analytic approach emphasized data reduction (module construction) and cross-cohort replication to identify robust biological signals within each gene group while preserving statistical power; however, an increased sample size would support more detailed subgroup comparisons focused on individual proteins, such as candidate biomarkers that discriminate genetic FTLD-TDP versus FTLD-tau. Another limitation is that our discovery cohort proteomic search was restricted to targets captured by SomaScan, which may introduce bias in the pathways identified. SomaScan has been applied in other neurodegenerative diseases21,25,60,61; however, platforms such as Olink and TMT-MS may exhibit greater coverage for certain molecular pathways (for example, immune) and in the case of MS, the ability to more directly query the abundance of protein isoforms and post-translational modifications (for example, phosphorylation). A recent study measured 1,981 CSF proteins via TMT-MS across familial FTLD mutation carriers, similarly identifying synaptic, immune and ECM co-expression modules linked to disease progression62. In contrast to the current study, this TMT-MS approach did not reveal an FTLD RNA metabolism proteomic signature, consistent with prior work demonstrating that SomaScan has broader coverage of RNA metabolism proteins21. Importantly, our cross-cohort/cross-platform/cross-disease analyses recapitulated SomaScan FTLD signatures with Olink data and demonstrated overlap between our SomaScan-only network and prior networks that also included TMT-MS. Moving forward, assaying the same FTLD patient samples across multiple platforms will be important for advancing clinical biomarker development of top aptamer-based targets identified during discovery.
Our analysis of over 4,000 CSF proteins in FTLD, to our knowledge, reflects the largest number of proteins measured in FTLD CSF so far. Our findings highlight a diverse ensemble of CSF protein network alterations in FTLD that align with preclinical knowledge of FTLD pathophysiology. We demonstrate that FTLD CSF proteomic signatures, including those probably reflecting alterations in RNA processing and synaptic biology, are robust and reproducible across genetic and sporadic FTLD cohorts. Our integrative systems-biology approach also highlights the potential for FTLD proteomics to inform pathophysiological mechanisms that are shared and distinct across proteinopathies, including mixed tauopathies such as AD. Importantly, this FTLD CSF proteomic network provides a data anchor that can be leveraged for multilayered analyses across matrices and omics pathways, thereby offering a rich resource for FTLD and neurodegenerative biomarker research.
Methods
The ALLFTD cohort
Participants included 116 carriers of pathogenic mutations in the C9orf72 (n = 47), GRN (n = 32) or MAPT (n = 37) genes and 39 non-carrier controls from families with a known mutation in one of these genes63. Participants were enrolled between 2015 and 2020 in the ALLFTD consortium (NCT04363684)10,64, which includes 23 collaborating centers across the USA and Canada. The ALLFTD study was approved through the Trial Innovation Network at Johns Hopkins University. Local ethics committees at each of the sites approved the study, and all participants provided written informed consent or assent with proxy consent. Inclusion in the present study required completion of baseline lumbar puncture for CSF collection. Four participants were excluded from final analysis due to outlier CSF samples (see below). Clinical stage was determined with the CDR + NACC-FTLD65,66,67. Mutation carriers were classified as presymptomatic or symptomatic based on a CDR + NACC-FTLD global score of 0 (presymptomatic) or ≥0.5 (symptomatic). Clinical and demographic characteristics across the three gene groups and controls are provided in Supplementary Table 1.
The 4RTNI cohort
A cohort of 35 patients with a clinical diagnosis of PSP-RS were recruited through the University of California, San Francisco (UCSF) Memory and Aging Center and the 4-Repeat Tauopathy Neuroimaging Initiative (4RTNI; NCT01804452). Of these 35 patients, 20 went to autopsy at the UCSF Memory and Aging Center Neurodegenerative Disease Brain Bank and had a confirmatory primary pathological diagnosis of PSP. A comparison group of 30 community-dwelling, cognitively unimpaired controls were recruited through the UCSF Brain Aging Network for Cognitive Health. CSF was collected on all participants. Research procedures were approved by the UCSF Institutional Review Board and all participants provided written informed consent or assent with proxy consent.
CSF collection and processing
CSF was collected at baseline in polypropylene tubes via lumbar puncture in lateral recumbent or sitting positions. CSF samples were centrifuged at 2,000g for 10 min at room temperature. Supernatant was aliquoted in 500 µl polypropylene tubes and stored at −80 °C until further analyses. All CSF samples underwent only one freeze/thaw cycle before analysis. CSF ALLFTD samples were analyzed for NfL using Quanterix Simoa, as previously described4.
SOMAmer proteomics
CSF samples were analyzed using a proprietary version of the SomaScan proteomics platform (SomaLogic) that captured 4,138 unique proteins. Samples were randomized before analysis on single plates, with experimenters blinded to clinical information. The SomaScan.adat file was loaded using the SomaDataIO R package v6.2.0. The SomaScan assay leverages slow off-rate modified aptamers (SOMAmers), which are short, single-stranded deoxynucleotides that bind to protein targets with high specificity68. A volume of 65 µl of CSF was used to create SOMAmer–protein reactions in 96-well plates. Tagged SOMAmer–protein complexes were captured in a bead-based assay, and levels of SOMAmer bound to sample were quantified through a fluorescent signal in DNA hybridization microarrays69. The reaction signal was detected digitally and expressed as aggregated Agilent relative fluorescent units, which were normalized to scale and subsequently log2 transformed for downstream analysis.
Outlier removal and covariate adjustment
Each sample had full protein data availability with no missing values. Outlier samples were identified (n = 6 (2 C9orf72, 1 GRN and 1 control from the genetic cohort, 2 controls from the sporadic cohort)) and removed using a threefold s.d. cutoff of z-transformed sample connectivity, as previously described20. Nonparametric bootstrap regression was performed in controls (familial and non-familial) to obtain a median estimated coefficient from 1,000 iterations of fitting for the effects of typical aging and sex on each protein. Median estimated coefficients were subsequently multiplied by age and sex and subtracted from participant-specific protein values to derive adjusted protein values across the entire log2(abundance) matrix of cases and controls in ALLFTD and 4RTNI cohorts.
CSF protein co-expression network analysis
In the genetic cohort, the WGCNA algorithm20,70 was used to generate a CSF protein network from the n = 4,138 log2 protein abundance × n = 155 case–sample matrix that had undergone covariate correction and network connectivity outlier removal as described above. The WGCNA blockwiseModules function was run with the following parameters: power = 12, deepSplit = 4, minModuleSize = 30, mergeCutHeight = 0.07, TOMdenom = ‘mean’, bicor correlation, signed network type, PAM staging and PAM respects dendro as TRUE, with clustering completed within a single block. We specified networkType = ‘signed’ in WGCNA, which preserves the sign of correlations by considering negatively correlated protein pairs as unconnected rather than connected with positive strength. While the adjacency matrix remains non-negative by design, this approach maintains directional relationships and prevents the merging of oppositely regulated proteins. To balance sparsity and biological interpretability, we selected a soft-thresholding power of 12 that surpassed a scale-free topology fit index (R2) of 0.80 (0.81) and exhibited mean network connectivity <100 (14.03). Module memberships were then iteratively reassigned to enforce kME table consistency, as previously described20. This module membership reassignment procedure increased the smallest module size in the network from 34 to 48 (M31) and reduced the gray (unassigned) protein count for the network from 832 (20.1%) to 199 (4.8%). Hub proteins were defined as proteins in the top 20th percentile of kME within each module20,71, enabling identification of multiple centrally connected proteins per module consistent with the distributed organization of biological networks72.
GO and cell type enrichment
To characterize protein module biology, we retrieved GO annotations from the Bader Lab’s monthly updated.GMT formatted ontology lists downloaded 2 October 202373. A Fisher’s exact test for enrichment was performed for each module’s protein membership with the background proteome consisting of all proteins measured in the current study. The pruned output with overrepresentation z scores was visualized using an R script. Modules were also tested for cell type-specific enrichment with Fisher’s exact one-tailed tests that compared module protein members with an in-house human cell type marker list20,74.
Differential abundance analysis
Differentially abundant proteins were identified by one-way ANOVA, followed by Tukey’s post hoc correction for pairwise comparisons between symptomatic carriers, presymptomatic carriers, and controls (stratified analysis for each gene). Volcano plots were made using an R script that color-coded individual data points by CSF network module membership.
Cognitive trajectory analysis
ALLFTD participants completed an annual comprehensive neuropsychological battery, covering episodic memory, executive functions and language skills (average (range) visits of 3.4 (1–7))10,75. Raw test scores were converted to z scores based on the score distribution in the larger cognitively unimpaired UCSF Brain Aging Network for Cognitive Health cohort76, which were then averaged into domain-based composite scores. Domain-based composite scores were averaged together to create a global cognitive composite score. The rate of cognitive change over time for each participant was determined via a linear mixed-effects model that examined global cognitive scores as a function of time (years since baseline), entered as a fixed and random effect, adjusting for baseline age, sex and years of education. Person-specific random slopes were extracted to represent the annual rate of cognitive change per participant, which was the primary outcome in downstream analyses that examined differential correlation of modules and individual proteins with cognitive trajectory.
BioFINDER 2 cohort
To test whether FTLD CSF proteomic signatures were robust to platform and cohorts, we leveraged CSF proteomic data analyzed on the Olink platform in the Swedish BioFINDER 2 cohort (NCT03174938)77. This BioFINDER 2 replication cohort included 29 patients with frontotemporal dementia clinical syndromes, 87 AD biomarker-positive patients with MCI or dementia, and 248 AD biomarker-negative controls78. Of the 29 FTLD patients, 10 had genetically confirmed FTLD (8 C9orf72, 1 GRN, 1 MAPT) with a behavioral variant frontotemporal dementia clinical syndrome. The remaining 19 cases included 13 behavioral variant frontotemporal dementia and 6 semantic variant primary progressive aphasia. AD cases were matched to FTLD spectrum cases three-to-one for age, sex, disease severity and mean protein level77 using k-nearest neighbors. BioFINDER 2 participants were recruited at Skåne University Hospital and the Hospital of Ängelholm, Sweden, diagnosed by multidisciplinary assessment after clinical and neuropsychological examination, brain magnetic resonance imaging and lumbar puncture. The study was approved by the Regional Ethics Committee in Lund, Sweden. All participants gave written informed consent to participate.
Module preservation validation analysis
A sporadic PSP-RS (4RTNI) protein co-expression network (4,138 log2 protein abundance × 74 case–sample matrix) was constructed using the same blockwiseModules parameters and module reassignment procedures as described above. Preservation between the genetic FTLD and sporadic PSP-RS CSF networks was performed using the WGCNA modulePreservation function with 500 permutations. Zsummary composite preservation scores for eight underlying network parameters were obtained using the mutation network as the template and PSP-RS network as the target20,79. Statistical significance of module preservation (minus log10 (FDR-adjusted P values)) was visualized as a function of module size.
Synthetic modules validation analysis
Network validation was also assessed in the 4RTNI and BioFINDER 2 cohorts by calculating synthetic eigenproteins20,22, which captures the variance of all protein module members present in the target cohorts. We leveraged every protein module membership assignment from the genetic FTLD network to construct synthetic eigenproteins in sporadic PSP-RS, given that CSF from both cohorts were analyzed on the same platform. Synthetic eigenprotein calculations in the BioFINDER 2 Olink dataset were restricted to Olink proteins that overlapped with SomaScan proteins from the genetic FTLD network and had missing frequency <75% (856 total proteins). Box plots visualized synthetic eigenprotein differences across cases and controls in each target cohort.
Correlation with AD and PD datasets validation analysis
Using recently published CSF SomaScan datasets, we correlated gene-stratified symptomatic FTLD versus control log2-fold change effect sizes to AD versus control effect sizes (reported by Dammer et al.27) and PD versus control effect sizes (reported by Rutledge et al.28). Correlations were conducted for the subset of differentially abundant proteins in symptomatic FTLD versus control comparisons that had exact SomaScan aptamer matches in the external datasets. Module-wise overlap with AD- and PD-associated proteins was determined using one-tailed Fisher’s exact test, followed by correction of P values for multiple testing using the Benjamini–Hochberg method.
Module overrepresentation validation analysis
To determine the degree to which CSF modules in the genetic FTLD network overlapped with CSF modules in AD, we performed overrepresentation analysis (ORA) of module gene symbols between the CSF genetic FTLD network and two CSF AD protein co-expression networks that differed in depth of proteomic coverage and platform. Specifically, the CSF genetic FTLD network was compared with (1) a multiplatform CSF network (Somalogic, TMT-MS; n = 5,242 proteins, 34 modules) constructed in a small cohort of AD (n = 160) and controls (n = 140)27 and (2) a shallower TMT-MS CSF network (n = 1,840 proteins, 14 modules) constructed in a more racially diverse (~50% African American) cohort of AD (n = 98) and controls (n = 105)30. Module overlap was determined using a one-tailed Fisher’s exact test, followed by correction of P values for multiple testing using the Benjamini–Hochberg method. Overlap of module gene symbols between networks was visualized using a custom in-house script27.
Statistics and reproducibility
Study sample sizes were determined by biofluid sample availability rather than an a priori statistical method. Statistical analyses were performed in R (v4.3.1). Box plots represent the median, 25th, and 75th percentile extremes; thus, hinges of a box represent the interquartile range of the two middle quartiles of data within a group. Minimum and maximum data points define the extent of whiskers (error bars). We visually inspected the distribution of module eigenproteins and individual proteins to confirm normality, but no formal statistical tests were performed. Correlations were performed using Spearman’s rho (ρ) coefficients. Comparisons between two groups were performed by a two-sided t-test. Comparisons among three or more groups were performed with ANOVA with Tukey’s pairwise comparison of significance. P values were adjusted for multiple comparisons by FDR correction according to the Benjamini–Hochberg method where indicated.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
Raw data are available upon reasonable request with formal applications to each respective study. ALLFTD and 4RTNI data requests can be completed via ALLFTD at https://www.allftd.org/data. Data from the UCSF Memory and Aging Center can be similarly requested via UCSF at https://memory.ucsf.edu/research-trials/professional/open-science. Certain data elements may be restricted due to the potential for identifiability in the context of the sensitive nature of the genetic data. For BioFINDER 2 data, anonymized data will be shared by request from a qualified academic investigator as long as data transfer is in agreement with European Union legislation on the general data protection regulation and decisions by the Ethical Review Board of Sweden and Region Skåne, which should be regulated in a material transfer agreement (BioFINDER 2 data requests addressed to bf_executive@med.lu.se). Publicly available summary statistics from Dammer et al.27 and Rutledge et al.28 can be accessed via the supplementary materials for each respective publication. Any other data reported in this paper are available from the lead contact upon reasonable request.
Code availability
Algorithms used for protein data processing and analysis are available in existing R packages and via GitHub at https://github.com/edammer, as described in the Methods.
Change history
07 July 2025
A Correction to this paper has been published: https://doi.org/10.1038/s43587-025-00931-0
References
Bang, J., Spina, S. & Miller, B. L. Frontotemporal dementia. Lancet 386, 1672–1682 (2015).
Knopman, D. S. & Roberts, R. O. Estimating the number of persons with frontotemporal lobar degeneration in the US population. J. Mol. Neurosci. 45, 330–335 (2011).
Mackenzie, I. R. A. & Neumann, M. Molecular neuropathology of frontotemporal dementia: insights into disease mechanisms from postmortem studies. J. Neurochem. 138, 54–70 (2016).
Rojas, J. C. et al. Plasma neurofilament light for prediction of disease progression in familial frontotemporal lobar degeneration. Neurology 96, e2296–e2312 (2021).
Rosen, H. J., Boeve, B. F. & Boxer, A. L. Tracking disease progression in familial and sporadic frontotemporal lobar degeneration: recent findings from ARTFL and LEFFTDS. Alzheimers Dement. 16, 71–78 (2020).
Del Campo, M. et al. New developments of biofluid-based biomarkers for routine diagnosis and disease trajectories in frontotemporal dementia. Alzheimers Dement. 18, 2292–2307 (2022).
Gifford, A., Praschan, N., Newhouse, A. & Chemali, Z. Biomarkers in frontotemporal dementia: current landscape and future directions. Biomark. Neuropsychiatry 8, 100065 (2023).
Boxer, A. L. et al. New directions in clinical trials for frontotemporal lobar degeneration: methods and outcome measures. Alzheimers Dement. 16, 131–143 (2019).
Greaves, C. V. & Rohrer, J. D. An update on genetic frontotemporal dementia. J. Neurol. 266, 2075–2086 (2019).
Staffaroni, A. M. et al. Temporal order of clinical and biomarker changes in familial frontotemporal dementia. Nat. Med. 28, 2194–2206 (2022).
Ferrari, R., Manzoni, C. & Hardy, J. Genetics and molecular mechanisms of frontotemporal lobar degeneration: an update and future avenues. Neurobiol. Aging 78, 98–110 (2019).
Balendra, R. & Isaacs, A. M. C9orf72-mediated ALS and FTD: multiple pathways to disease. Nat. Rev. Neurol. 14, 544–558 (2018).
Cook, C. N. et al. C9orf72 poly(GR) aggregation induces TDP-43 proteinopathy. Sci. Transl. Med. 12, eabb3774 (2020).
Kao, A. W., McKay, A., Singh, P. P., Brunet, A. & Huang, E. J. Progranulin, lysosomal regulation and neurodegenerative disease. Nat. Rev. Neurosci. 18, 325–333 (2017).
Kumar-Singh, S. Progranulin and TDP-43: mechanistic links and future directions. J. Mol. Neurosci. 45, 561–573 (2011).
Strang, K. H., Golde, T. E. & Giasson, B. I. MAPT mutations, tauopathy, and mechanisms of neurodegeneration. Lab. Invest. 99, 912–928 (2019).
Schmidt, S., Holzer, M., Arendt, T., Sonntag, M. & Morawski, M. Tau protein modulates perineuronal extracellular matrix expression in the TauP301L-acan mouse model. Biomolecules 12, 505 (2022).
Tracy, T. E. et al. Tau interactome maps synaptic and mitochondrial processes associated with neurodegeneration. Cell 185, 712–728.e714 (2022).
Rayaprolu, S. et al. Systems-based proteomics to resolve the biology of Alzheimer’s disease beyond amyloid and tau. Neuropsychopharmacology 46, 98–115 (2021).
Johnson, E. C. B. et al. Large-scale deep multi-layer analysis of Alzheimer’s disease brain reveals strong proteomic disease-related changes not observed at the RNA level. Nat. Neurosci. 25, 213–225 (2022).
Dammer, E. B. et al. Multi-platform proteomic analysis of Alzheimer’s disease cerebrospinal fluid and plasma reveals network biomarkers associated with proteostasis and the matrisome. Alzheimers Res. Ther. 14, 174 (2022).
Johnson, E. C. B. et al. Large-scale proteomic analysis of Alzheimer’s disease brain and cerebrospinal fluid reveals early changes in energy metabolism associated with microglia and astrocyte activation. Nat. Med. 26, 769–780 (2020).
Gendron, T. F. et al. Comprehensive cross-sectional and longitudinal analyses of plasma neurofilament light across FTD spectrum disorders. Cell Rep. Med. 3, 100607 (2022).
Delaby, C. et al. Differential levels of neurofilament light protein in cerebrospinal fluid in patients with a wide range of neurodegenerative disorders. Sci. Rep. 10, 9161 (2020).
Phillips, B. et al. Proteome wide association studies of LRRK2 variants identify novel causal and druggable proteins for Parkinson’s disease. npj Parkinsons Dis. 9, 107 (2023).
Coyle-Gilchrist, I. T. et al. Prevalence, characteristics, and survival of frontotemporal lobar degeneration syndromes. Neurology 86, 1736–1743 (2016).
Dammer, E. B. et al. Proteomic analysis of Alzheimer’s disease cerebrospinal fluid reveals alterations associated with APOE ε4 and atomoxetine treatment. Sci. Transl. Med. 16, eadn3504 (2024).
Rutledge, J. et al. Comprehensive proteomics of CSF, plasma, and urine identify DDC and other biomarkers of early Parkinson’s disease. Acta Neuropathol. 147, 52 (2024).
Trautwig, A. N. et al. Network analysis of the cerebrospinal fluid proteome reveals shared and unique differences between sporadic and familial forms of amyotrophic lateral sclerosis. Preprint at bioRxiv https://doi.org/10.1101/2024.02.29.582840 (2024).
Modeste, E. S. et al. Quantitative proteomics of cerebrospinal fluid from African Americans and Caucasians reveals shared and divergent changes in Alzheimer’s disease. Mol. Neurodegener. 18, 48 (2023).
Umoh, M. E. et al. A proteomic network approach across the ALS-FTD disease spectrum resolves clinical phenotypes and genetic vulnerability in human brain. EMBO Mol. Med. 10, 48–62 (2018).
Johnson, E. C. B. et al. Deep proteomic network analysis of Alzheimer’s disease brain reveals alterations in RNA binding proteins and RNA splicing associated with disease. Mol. Neurodegener. 13, 52 (2018).
Nikom, D. & Zheng, S. Alternative splicing in neurodegenerative disease and the promise of RNA therapies. Nat. Rev. Neurosci. 24, 457–473 (2023).
Liu, E. Y. et al. Loss of nuclear TDP-43 is associated with decondensation of LINE retrotransposons. Cell Rep. 27, 1409–1421.e1406 (2019).
Hofmann, J. W., Seeley, W. W. & Huang, E. J. RNA binding proteins and the pathogenesis of frontotemporal lobar degeneration. Annu Rev. Pathol. 14, 469–495 (2019).
Bampton, A., Gittings, L. M., Fratta, P., Lashley, T. & Gatt, A. The role of hnRNPs in frontotemporal dementia and amyotrophic lateral sclerosis. Acta Neuropathol. 140, 599–623 (2020).
Deshaies, J.-E. et al. TDP-43 regulates the alternative splicing of hnRNP A1 to yield an aggregation-prone variant in amyotrophic lateral sclerosis. Brain 141, 1320–1333 (2018).
Lester, E. & Parker, R. Tau, RNA, and RNA-binding proteins: complex interactions in health and neurodegenerative diseases. Neuroscientist 30, 458–472 (2023).
Irwin, K. E. et al. A fluid biomarker reveals loss of TDP-43 splicing repression in presymptomatic ALS–FTD. Nat. Med. 30, 382–393 (2024).
Chatterjee, M. et al. Plasma extracellular vesicle tau and TDP-43 as diagnostic biomarkers in FTD and ALS. Nat. Med. 30, 1771–1783 (2024).
Casaletto, K. B. et al. Neurogranin, a synaptic protein, is associated with memory independent of Alzheimer biomarkers. Neurology 89, 1782–1788 (2017).
Wingo, A. P. et al. Large-scale proteomic analysis of human brain identifies proteins associated with cognitive trajectory in advanced age. Nat. Commun. 10, 1619 (2019).
Camporesi, E. et al. Fluid biomarkers for synaptic dysfunction and loss. Biomark. Insights 15, 1177271920950319 (2020).
Paterson, R. W. et al. SILK studies—capturing the turnover of proteins linked to neurodegenerative diseases. Nat. Rev. Neurol. 15, 419–427 (2019).
Boiten, W. A. et al. Pathologically decreased CSF levels of synaptic marker NPTX2 in DLB are correlated with levels of alpha-synuclein and VGF. Cells 10, 38 (2020).
Libiger, O. et al. Longitudinal CSF proteomics identifies NPTX2 as a prognostic biomarker of Alzheimer’s disease. Alzheimers Dement. 17, 1976–1987 (2021).
van der Ende, E. L. et al. Neuronal pentraxin 2: a synapse-derived CSF biomarker in genetic frontotemporal dementia. J. Neurol. Neurosurg. Psychiatry 91, 612–621 (2020).
Weißflog, L. et al. KCNIP4 as a candidate gene for personality disorders and adult ADHD. Eur. Neuropsychopharmacol. 23, 436–447 (2013).
Tam, G. W. et al. Confirmed rare copy number variants implicate novel genes in schizophrenia. Biochem. Soc. Trans. 38, 445–451 (2010).
Zhang, L. et al. Network connectivity alterations across the MAPT mutation clinical spectrum. Ann. Neurol. 94, 632–646 (2023).
Lee, S. E. et al. Network degeneration and dysfunction in presymptomatic C9ORF72 expansion carriers. Neuroimage Clin. 14, 286–297 (2017).
Huber, N. et al. Deficient neurotransmitter systems and synaptic function in frontotemporal lobar degeneration—insights into disease mechanisms and current therapeutic approaches. Mol. Psychiatry 27, 1300–1309 (2022).
Hou, P. S., hAilín, D., Vogel, T. & Hanashima, C. Transcription and beyond: delineating FOXG1 function in cortical development and disorders. Front. Cell Neurosci. 14, 35 (2020).
Pitale, P. M., Howse, W. & Gorbatyuk, M. Neuronatin protein in health and disease. J. Cell. Physiol. 232, 477–481 (2017).
Pintér, P. & Alpár, A. The role of extracellular matrix in human neurodegenerative diseases. Int. J. Mol. Sci. 23, 11085 (2022).
De Luca, C., Colangelo, A. M., Virtuoso, A., Alberghina, L. & Papa, M. Neurons, glia, extracellular matrix and neurovascular unit: a systems biology approach to the complexity of synaptic plasticity in health and disease. Int. J. Mol. Sci. 21, 1539 (2020).
Johnson, E. C. B. et al. Cerebrospinal fluid proteomics define the natural history of autosomal dominant Alzheimer’s disease. Nat. Med. 29, 1979–1988 (2023).
Moretto, E., Stuart, S., Surana, S., Vargas, J. N. S. & Schiavo, G. The role of extracellular matrix components in the spreading of pathological protein aggregates. Front. Cell. Neurosci. 16, 844211 (2022).
Holmes, B. B. & Diamond, M. I. Prion-like properties of Tau protein: the importance of extracellular Tau as a therapeutic target. J. Biol. Chem. 289, 19855–19861 (2014).
Shi, L. et al. Discovery and validation of plasma proteomic biomarkers relating to brain amyloid burden by SOMAscan assay. Alzheimers Dement. 15, 1478–1488 (2019).
Yang, C. et al. Genomic atlas of the proteome from brain, CSF and plasma prioritizes proteins implicated in neurological disorders. Nat. Neurosci. 24, 1302–1312 (2021).
Sogorb-Esteve, A. et al. Proteomic analysis reveals distinct cerebrospinal fluid signatures across genetic frontotemporal dementia subtypes. Sci. Transl. Med. 17, eadm9654 (2025).
Ramos, E. M. et al. Genetic screening of a large series of North American sporadic and familial frontotemporal dementia cases. Alzheimers Dement. 16, 118–130 (2020).
Boeve, B. et al. The longitudinal evaluation of familial frontotemporal dementia subjects protocol: framework and methodology. Alzheimers Dement. 16, 22–36 (2020).
Knopman, D. S., Weintraub, S. & Pankratz, V. S. Language and behavior domains enhance the value of the clinical dementia rating scale. Alzheimers Dement. 7, 293–299 (2011).
Knopman, D. S. et al. Development of methodology for conducting clinical trials in frontotemporal lobar degeneration. Brain 131, 2957–2968 (2008).
Miyagawa, T. et al. Utility of the global CDR plus NACC FTLD rating and development of scoring rules: data from the ARTFL/LEFFTDS Consortium. Alzheimers Dement. 16, 106–117 (2020).
Rohloff, J. C. et al. Nucleic acid ligands with protein-like side chains: modified aptamers and their use as diagnostic and therapeutic agents. Mol. Ther. Nucleic Acids 3, e201 (2014).
Gold, L. et al. Aptamer-based multiplexed proteomic technology for biomarker discovery. PLoS ONE 5, e15004 (2010).
Langfelder, P. & Horvath, S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics 9, 559 (2008).
Shantaraman, A. et al. Network proteomics of the Lewy body dementia brain reveals presynaptic signatures distinct from Alzheimer’s disease. Mol. Neurodegener. 19, 60 (2024).
Langfelder, P., Mischel, P. S. & Horvath, S. When is hub gene selection better than standard meta-analysis? PLoS ONE 8, e61505 (2013).
Reimand, J. et al. Pathway enrichment analysis and visualization of omics data using g:Profiler, GSEA, Cytoscape and EnrichmentMap. Nat. Protoc. 14, 482–517 (2019).
Zhang, Y. et al. An RNA-sequencing transcriptome and splicing database of glia, neurons, and vascular cells of the cerebral cortex. J. Neurosci. 34, 11929–11947 (2014).
Weintraub, S. et al. Version 3 of the Alzheimer Disease Centers’ neuropsychological test battery in the uniform data set (UDS). Alzheimer Dis. Assoc. Disord. 32, 10–17 (2018).
Saloner, R. et al. Plasma phosphorylated tau-217 exhibits sex-specific prognostication of cognitive decline and brain atrophy in cognitively unimpaired adults. Alzheimers Dement. 20, 376–387 (2024).
Karlsson, L. et al. Cerebrospinal fluid reference proteins increase accuracy and interpretability of biomarkers for brain diseases. Nat. Commun. 15, 3676 (2024).
Santillo, A. F. et al. [18F]RO948 tau positron emission tomography in genetic and sporadic frontotemporal dementia syndromes. Eur. J. Nucl. Med. Mol. Imaging 50, 1371–1383 (2023).
Langfelder, P., Luo, R., Oldham, M. C. & Horvath, S. Is my network module preserved and reproducible? PLoS Comput. Biol. 7, e1001057 (2011).
Acknowledgements
Data collection and dissemination of the data presented in this paper were supported by the ALLFTD Consortium (U19: AG063911 (A.L.B., H.J.R. and B.F.B.), funded by the National Institute on Aging and the National Institute of Neurological Diseases and Stroke) and the former ARTFL and LEFFTDS Consortia (ARTFL: U54 NS092089 (A.L.B., H.J.R. and B.F.B.), funded by the National Institute of Neurological Diseases and Stroke and National Center for Advancing Translational Sciences; LEFFTDS (U01 AG045390 (H.J.R. and B.F.B.)), funded by the National Institute on Aging and the National Institute of Neurological Diseases and Stroke). The paper was reviewed by the ALLFTD Executive Committee for scientific content. We acknowledge the invaluable contributions of the study participants and families as well as the assistance of the support staff at each of the participating sites. This work is also supported by the Alzheimer’s Association (grant nos. AARF-23-1145318 (R.S.), AARF-22-974065 (E.W.P.) and AARG-20-683875 (K.B.C.)), New Vision Research (grant no. CCAD 2024-001-1 (R.S.)), an American Academy of Neurology and Association for Frontotemporal Degeneration Fellowship Award (R.S.), the Bluefield Project to Cure FTD, CurePSP, Larry L. Hillblom Foundation (grant nos. 2018-A-006-NET (J.H.K.) and 2018-A-025-FEL (A.M.S.)), the National Institutes of Health (grant nos. AG038791 (A.L.B.), AG032306 (H.J.R.), AG016976 (W.K.), K23AG059888 (J.C.R.), K23AG073514 (L.V.), AG019724 (B.L.M.), AG058233 (S.E.L.), P30 AG062422 (W.W.S.), R01AG072475 (PI: K.B.C.), K23AG061253 (A.M.S.), R01AG032289 (J.H.K.), R01AG048234 (J.H.K.) and K24AG045333 (H.J.R.)) and the Rainwater Charitable Foundation. Samples from the National Centralized Repository for Alzheimer Disease and Related Dementias, which receives government support under a cooperative agreement grant (no. U24 AG021886) awarded by the National Institute on Aging (NIA), were used in this study. Work at Lund University was supported by the GHR Foundation, Swedish Research Council (2022-00775), ERA PerMed (ERAPERMED2021-184), Knut and Alice Wallenberg foundation (2022-0231), Strategic Research Area MultiPark (Multidisciplinary Research in Parkinson’s disease) at Lund University, Swedish Alzheimer Foundation (AF-980907), Swedish Brain Foundation (FO2021-0293), Parkinson foundation of Sweden (1412/22), Cure Alzheimer’s fund, Rönström Family Foundation, Konung Gustaf V:s och Drottning Victorias Frimurarestiftelse, Skåne University Hospital Foundation (2020-O000028), Regionalt Forskningsstöd (2022-1259) and Swedish federal government under the ALF agreement (2022-Projekt0080). The funding sources had no role in the design and conduct of the study; in the collection, analysis, interpretation of the data; or in the preparation, review or approval of the paper.
Author information
Authors and Affiliations
Consortia
Contributions
R.S. had full access to the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. R.S., J.C.R., A.L.B. and K.B.C. were responsible for concept development and design. R.S. and J.W.V. conducted statistical analyses. E.B.D. developed and maintained the code repositories used for proteomic data analysis and visualization. R.S. drafted the paper. A.M.S., J.C.R., A.L.B. and K.B.C. critically revised the paper. A.L.B. supervised the research. O.H., B.F.B., H.J.R., J.C.R. and A.L.B. obtained funding. All authors contributed to acquisition, analysis or interpretation of data or revision of the paper.
Corresponding author
Ethics declarations
Competing interests
A.M.S. received research support from the NIA/NIH, the Bluefield Project to Cure FTD, and the Larry L. Hillblom Foundation. He has provided consultation to Alector, Lilly, Passage Bio and Takeda. L.K.F. receives research support from the NIH. O.H. has received research support (for the institution) from AVID Radiopharmaceuticals, Biogen, C2N Diagnostics, Eli Lilly, Eisai, Fujirebio, GE Healthcare and Roche. In the past 2 years, he has received consultancy/speaker fees from AC Immune, Alzpath, BioArctic, Biogen, Bristol Meyer Squibb, Cerveau, Eisai, Eli Lilly, Fujirebio, Merck, Novartis, Novo Nordisk, Roche, Sanofi and Siemens. J.S.Y. receives research support from the NIH. P.A.L. is a site primary investigator for clinical trials by Alector, AbbVie and Woolsey. He serves as an advisor for Retrotrope. He receives research and salary support from the NIH–NIA and the Alzheimer’s Association-Part the Cloud partnership. E.B. receives research support from the NIH and Lewy Body Dementia Association. B.C.D. is a consultant for Acadia, Alector, Arkuda, Biogen, Denali, Eisai, Genentech, Lilly, Merck, Novartis, Takeda and Wave Lifesciences; receives royalties from Cambridge University Press, Elsevier and Oxford University Press; and receives grant funding from the NIA, the National Institute of Neurological Disorders and Stroke, the National Institute of Mental Health and the Bluefield Foundation. K.D.-R. receives research support from the NIH and serves as an investigator for a clinical trial sponsored by Lawson Health Research Institute. K.F. receives research support from the NIH. N.G. has participated or is currently participating in clinical trials of anti-dementia drugs sponsored by Bristol Myers Squibb, Eli Lilly/Avid Radiopharmaceuticals, Janssen Immunotherapy, Novartis, Pfizer, Wyeth, SNIFF (The Study of Nasal Insulin to Fight Forgetfulness) and the A4 (The Anti-Amyloid Treatment in Asymptomatic Alzheimer’s Disease) trial; and receives research support from Tau Consortium, the Association for Frontotemporal Dementia and the NIH. N.G.-R. receives royalties from UpToDate and has participated in multicenter therapy studies by sponsored by Biogen, TauRx, and Lilly; and receives research support from the NIH. C.M.H. is a Site PI or SubI for several industry (Alector, Janssen, Biogen, Cogito Tx) sponsored clinical trials with funding through Emory Office of Sponsored Programs. L.S.H. receives research funding from Abbvie, Acumen, Alector, Biogen, BMS, Eisai, Genentech/Roche, Janssen/J&J, Transposon, UCB, Vaccinex; and receives consulting fees from Biogen, Cortexyme, Eisai, Medscape, Prevail/Lilly. G.-Y.R.H. has served as an investigator for clinical trials sponsored by AstraZeneca, Eli Lilly and Roche/Genentech; and he receives research support from the Canadian Institutes of Health Research and the Alzheimer Society of British Columbia. E.D.H. receives research support from the NIH. J. Kornak has provided expert witness testimony for Teva Pharmaceuticals in Forest Laboratories Inc. et al. v Teva Pharmaceuticals USA, Inc., case numbers 1:14-cv-00121 and 1:14-cv-00686 (D. Del. filed 31 January 2014 and 30 May 2014 regarding the drug Memantine) and for Apotex/HEC/Ezra in Novartis AG et al. v Apotex Inc., case number 1:15-cv-975 (D. Del. filed 26 October 2015 regarding the drug Fingolimod); he has also given testimony on behalf of Puma Biotechnology in Hsingching Hsu et al. v Puma Biotechnology, Inc., et al. 2018 regarding the drug neratinib; and he receives research support from the NIH. W.K. receives research funding from AstraZeneca, Biogen, Roche, the Department of Defense and the NIH. M.I.L. receives research support from the NIH. The research of I.L. is supported by the National Institutes of Health grants 2R01AG038791-06A, U01NS100610, U01NS80818, R25NS098999; U19 AG063911-1 and 1R21NS114764-01A1; the Michael J Fox Foundation, Parkinson Foundation, Lewy Body Association, CurePSP, Roche, Abbvie, Biogen, Centogene. EIP-Pharma, Biohaven Pharmaceuticals, Novartis, Brain Neurotherapy Bio and United Biopharma SRL–UCB. She is a scientific advisor for Amydis and Rossy Center for Progressive Supranuclear Palsy University of Toronto. She receives her salary from the University of California San Diego and as Chief Editor of Frontiers in Neurology. C.T.M. receives funding from NIH and Penn Institute on Aging. M.F.M. receives research support from the NIH. A.P. receives research support from the NIH (U01 NS102035; K23 AG059891). H.L.P. receives research support from the NIH. L.P. receives research support from the NIH. E.M.R. receives research support from the NIH. K.R. receives research support from the NIH. E.D.R. has received research support from the NIH, the Bluefield Project to Cure Frontotemporal Dementia, the Alzheimer’s Association, the Alzheimer’s Drug Discovery Foundation, the BrightFocus Foundation, and Alector; has served as a consultant for AGTC and on a data monitoring committee for Lilly; and owns intellectual property related to tau and progranulin. R.S. receives support from the NIA, the National Institute of Neurological Disorders and Stroke, the Parkinson’s Disease Foundation and Acadia Pharmaceuticals. M.C.T. has served as an investigator for clinical trials sponsored by Biogen, Avanex, Green Valley, Roche/Genentech, Bristol Myers Squibb, Eli Lilly/Avid Radiopharmaceuticals and Janssen. She receives research support from the Canadian Institutes of Health Research. B.F.B. has served as an investigator for clinical trials sponsored by Alector, Biogen, Transposon and Cognition Therapeutics. He serves on the Scientific Advisory Board of the Tau Consortium which is funded by the Rainwater Charitable Foundation. He receives research support from NIH. H.J.R. has received research support from Biogen Pharmaceuticals, has consulting agreements with Wave Neuroscience, Ionis Pharmaceuticals, Eisai Pharmaceuticals and Genentech, and receives research support from the NIH and the state of California. J.C.R. receives research support from the NIH and is a site principal investigator for clinical trials sponsored by Eli Lilly and Eisai. A.L.B. receives research support from the NIH, the Tau Research Consortium, the Association for Frontotemporal Degeneration, Bluefield Project to Cure Frontotemporal Dementia, Corticobasal Degeneration Solutions, the Alzheimer’s Drug Discovery Foundation and the Alzheimer’s Association. He has served as a consultant for Aeovian, AGTC, Alector, Arkuda, Arvinas, Boehringer Ingelheim, Denali, GSK, Life Edit, Humana, Oligomerix, Oscotec, Roche, TrueBinding, Wave, Merck and received research support from Biogen, Eisai and Regeneron. The other authors declare no competing interests.
Peer review
Peer review information
Nature Aging thanks Emily Feneberg and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data
Extended Data Fig. 1 Differential Abundance.
Gene-stratified differential abundance analyses examined individual protein differences across symptomatic (Sx) carriers, presymptomatic (PreSx) carriers, and familial non-carrier controls (CTL). Pairwise differential protein abundance is represented by volcano plots of average log2 difference by negative log10 p-value for each given comparison. Proteins are colored by the module in which they were assigned according to the heatmap shown in Fig. 2a. Annotated proteins represent the top 10 differentially abundant proteins by statistical significance for each direction. Pairwise comparisons were performed using one-way ANOVA with Tukey test. The red horizontal dashed line represents Tukey p < 0.05.
Extended Data Fig. 2 ROC Curves.
Gene-stratified receiver-operating characteristic (ROC) analyses tested the discriminative utility of differentially abundant proteins across symptomatic (Sx) carriers, presymptomatic (PreSx) carriers, and familial non-carrier controls (CTL). ROC curves for the top 5 performing proteins, defined by area under the curve (AUC), are plotted alongside a curve derived from a combined panel of the top 5 proteins.
Extended Data Fig. 3 Correlation of Genetic FTLD-associated Protein Effect Sizes with External AD and PD Datasets.
a-f) Log2-fold changes of CSF SomaScan proteins with differential abundance in symptomatic FTLD gene carriers versus control (CTL) were correlated against CSF SomaScan protein effect sizes for AD versus CTL (A, C, E; log2-fold changes reported in Dammer et al.27) and PD versus CTL (B, D, F; regression coefficient weights reported in Rutledge et al.28). Correlations were performed using Pearson correlation. g) Genetic FTLD network module protein overlap with proteins increaed or decreased in AD or PD datasets. One-tailed Fisher’s exact test was used to determine module-wise enrichment, and results FDR-corrected using the Benjamini-Hochberg method. The exact p-values are reported in Supplementary Table 11.
Supplementary information
Supplementary Information
ALLFTD Consortium member list and Supplementary Fig. 1.
43587_2025_878_MOESM3_ESM.xlsx
Supplementary Table 1. Genetic FTLD cohort characteristics. Supplementary Table 2. CSF network module memberships. Supplementary Table 3. CSF network module GO term enrichment. Supplementary Table 4. CSF network eigenprotein group differences. Supplementary Table 5. CSF network proteins differential abundance. Supplementary Table 6. Count (percentage) of differentially abundant proteins per module. Supplementary Table 7. ROC curve analyses. Supplementary Table 8. Module correlations with global cognitive trajectories. Supplementary Table 9. CSF network changes in sporadic PSP-RS. Supplementary Table 10. CSF network changes in clinical FTD and AD. Supplementary Table 11. Genetic FTLD module-wise enrichment of AD and PD proteins. Supplementary Table 12. Genetic FTLD CSF network overlap with AD CSF networks.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Saloner, R., Staffaroni, A.M., Dammer, E.B. et al. Large-scale network analysis of the cerebrospinal fluid proteome identifies molecular signatures of frontotemporal lobar degeneration. Nat Aging 5, 1143–1158 (2025). https://doi.org/10.1038/s43587-025-00878-2
Received:
Accepted:
Published:
Issue date:
DOI: https://doi.org/10.1038/s43587-025-00878-2
This article is cited by
-
Shared and disease-specific pathways in frontotemporal dementia and Alzheimer’s and Parkinson’s diseases
Nature Medicine (2025)
-
FTLD-associated proteomic signatures uncovered
Nature Reviews Neurology (2025)
